Geographic differences and preferential mappings

Many assessments have suggested that the current array of PAs is not based solely on ecological value, but fundamentally in relation to cultural and economic imperatives (Margules & Pressey Reference Margules and Pressey2000). In the traditions of environmental history, landscape studies and people-environment geography, it has been argued that landscapes preferentially selected for protection are typically those that are considered to be dramatic or of high recreational value (Vale Reference Vale1988; Zimmerer et al. Reference Zimmerer, Galt and Buck2004). In North America, landscapes widely deemed to be ‘sublime’ are favoured over other landscapes that may have comparable or greater ecological importance (Cronon Reference Cronon and Cronon1995). Related discussions highlight ways in which particular visions of ‘nature’ or ‘wilderness’, especially notions of pristine nature that underwrite many conservation strategies, are necessarily intertwined with social, economic, cultural and historical processes (Cronon Reference Cronon and Cronon1995; Braun & Castree Reference Braun and Castree1998). Notions of ‘naturalness’ are often defined according to particular historical benchmarks; for instance, an area may be considered ‘natural’ if it conforms to visions of a pre-industrialization landscape, despite growing awareness that such landscapes were also heavily modified by human activities (Ridder Reference Ridder2007, p. 9). On a global scale, expectations of naturalness and wilderness can be seen through an emphasis on the protection of widely-publicized conservation targets, such as tropical rainforests, over other biomes, like grasslands, that have not received such focused attention (Hazen & Anthamatten Reference Hazen and Anthamatten2004). Given such debates related to which visions of nature to ‘conserve’, and the way that these vary according to cultural and social attributes, conservation decisions are not value-free (Whittaker et al. Reference Whittaker, Araújo, Jepson, Ladle, Watson and Willis2005). In response, increasing efforts are being made within the conservation sciences to try to make future conservation decisions more systematically (Margules & Pressey Reference Margules and Pressey2000).

Addressing these issues from a cartographic approach, an issue that has not yet been explored in the literature is the degree to which areas may also be selected for protection owing to their relative ‘mappability’. For instance, grasslands are not only less ‘sublime’ than many other landscapes, but are also less definable in cartographic terms (less ‘mappable’) than clearly-bounded features such as lakes, islands or even forests. The case of marine ecosystems is noteworthy in this regard, with data limitations, mobile features and other social and political considerations contributing to the difficulty of mapping and managing ocean features (Steinberg Reference Steinberg2001), and perhaps helping to explain why marine ecosystems have not seen the same proliferation of PAs over the past 20 years that has occurred in terrestrial areas (Brooks et al. Reference Brooks, Mittermeier, de Fonseca, Gerlach, Hoffmann, Lamoreux, Mittermeier, Pilgrim and Rodrigues2006). Related to such ideas, issues of accessibility and data availability also pose challenges for conservation mapping efforts. For example, spatial sampling biases in the intensity of biodiversity data collection in sub-Saharan Africa, with more data available for sites near urban areas and in more accessible regions, presents the potential for spatial data availability and accuracy to influence priority-setting activities (Reddy & Dávalos Reference Reddy and Dávalos2003).

Other clear influences on conservation mappings relate to pre-existing political and administrative boundaries, often affecting how conservation boundaries are drawn, as well as contributing to the neglect of certain important ecosystems (Hanks Reference Hanks2000). Jurisdictional boundaries often define at least one edge of a PA. In such cases, political boundaries define PAs without regard for ecological conditions. Fagergren (Reference Fagergren, Knight and Landes1998) details the sort of complications that arise when trying to mediate this complexity of administrative boundaries, demonstrating the interplay between conservation mappings and other priorities. The boundaries of Big Cypress National Preserve in Florida were drawn to account not only for ecological goals, but also local township lines, lands suitable for a jetport, hunting camps, indigenous claims and private land ownership (Fagergren Reference Fagergren, Knight and Landes1998).

Extending our line of argument related to the relative ‘mappability’ of geographic features, it is easy to imagine ways that conservation practice might also be weighted preferentially towards those concepts and practices that are most readily expressed in map form, particularly as mapping practices hold cachet scientifically and politically (see Nietschmann Reference Nietschmann1995 for a discussion of the ‘power of maps’). In ‘biodiversity hotspots’, conservation attention is focused on limited areas that are considered to harbour very high levels of biodiversity as targets for protection (Myers et al. Reference Myers, Mittermeier, Mittermeier, Fonseca and Kent2000), and the approach's cartographic underpinnings are advantageous in terms of providing a visual and easily comprehensible way of displaying the scientific rationale to a lay audience. Given the limited geographic extent of the regions displayed, mapping also contributes to a sense that the idea is not only feasible, but even politically attractive in terms of maximizing returns on conservation investments. There is no doubt that this approach has been successful, with funding bodies supporting hotspot conservation measures to the tune of US$ 750 million (Brooks et al. Reference Brooks, Mittermeier, de Fonseca, Gerlach, Hoffmann, Lamoreux, Mittermeier, Pilgrim and Rodrigues2006). However, given that the hotspots approach may, for instance, turn attention from other important conservation targets and wider conservation goals (Kareiva & Marvier Reference Kareiva and Marvier2003), mapping may entrench support for particular visions of conservation, while at the same time neglecting alternative strategies.

Power inequalities in conservation mapping

Another theme that has been important to the work of geographers is consideration of the social and political implications of PA designation and management. The political ecology literature, in particular, details instances where certain social groups, livelihoods or forms of knowledge are excluded or marginalized by the creation of particular PAs, especially when the conservation boundary is strictly policed (for example Neumann Reference Neumann1992, Reference Neumann1995; Peluso Reference Peluso1993; Colchester Reference Colchester1996; Katz Reference Katz, Braun and Castree1998; Turner Reference Turner1999). For instance, unsubstantiated claims about pastoralists competing with wildlife in South-western Niger have resulted in the strict exclusion of local people from a PA, ironically exacerbating threats to wildlife (Turner Reference Turner1999). Connections between monitoring the boundaries of PAs and violence against local populations have also been highlighted (Peluso Reference Peluso1993). These and similar studies have underscored the need to examine assumptions about local populations in conservation planning (Agrawal & Gibson Reference Agrawal and Gibson1999; Weis Reference Weis2000).

A related discussion considers how different forms of knowledge are valued unequally, with important implications for the mapping, designation and management of conservation spaces. A critique common to these discussions is that Western or techno-scientific ideas are often treated preferentially over traditional or indigenous ways of knowing, even as traditional knowledge systems may involve complex understandings of ecosystem processes (Berkes et al. Reference Berkes, Kislaioglu, Folke and Gadgil1998). As just one example, Maasai knowledge has been excluded from conservation efforts in Northern Tanzania, further marginalizing the Maasai and discouraging more flexible land-use management practices to which local knowledge may be particularly well suited (Goldman Reference Goldman2003). Along similar lines, and highlighting our focus on mapping, there is also increasing attention to the possibility that technologies such as GIS exacerbate these trends by further cleaving distinctions between ‘expert’ and ‘lay’ or ‘indigenous’ knowledge. In response, a considerable literature has developed considering how alternative mapping practices can be revalued, and mapping democratized via notions such as participatory GIS (see Nietschmann Reference Nietschmann1995; Hodgson & Schroeder Reference Hodgson and Schroeder2002).

Related to our earlier discussion of why the current configuration of PAs has emerged rather than other possible arrangements, geographical approaches have also emphasized social and political inequalities that help explain these differences. For instance, concerns have been raised that neo-colonial relationships may be perpetuated by the actions of the large conservation organizations of the industrialized North as they operate in the global South (Chapin Reference Chapin2004). For instance, in eastern Africa, over 10% of territory in some countries is now managed by state and international organizations for conservation purposes (Neumann Reference Neumann1997). Neumann (Reference Neumann1997) also suggested that PAs can consolidate and extend state control over land, shifting control over resources away from local people. While this critique is valid, it is notable that when analysing the current map of PAs, the countries with the highest proportions of land protected are actually often in the industrialized North. However, this too can be seen as a reflection of how political structures, recreational opportunities or conservation awareness may be more influential in the determination of PAs than are biodiversity imperatives per se (see Zimmerer et al. Reference Zimmerer, Galt and Buck2004; Naughton-Treves et al. Reference Naughton-Treves, Holland and Brandon2005). It is also problematic given that 90% of annual conservation funding is estimated to originate from and be spent in economically rich countries (Brooks et al. Reference Brooks, Mittermeier, de Fonseca, Gerlach, Hoffmann, Lamoreux, Mittermeier, Pilgrim and Rodrigues2006), while most of the world's biodiversity is found in developing countries in need of donor support (Smith et al. Reference Smith, Muir, Walpole, Balmford and Leader-Williams2003). These examples add weight to the idea that political conditions, economic inequalities, histories of colonialism and uneven relations of power are fundamental to understanding the contemporary map of PAs.

Effectiveness and effects of mapped conservation spaces

Another broad area of enquiry from a geographic perspective relates to the effectiveness of, and effects of, PAs. These discussions evaluate the role that mapped conservation spaces play, their relative effectiveness for different conservation goals and other political, social or ecological effects that reliance on territorial conservation strategies may engender. Initiating these discussions, conservation practitioners have already detailed a number of ways in which protected areas are relatively ineffective. For instance, PA boundaries are often unable to prevent the entry of threats, whether pollution, invasive species or poachers (Woodley Reference Woodley, Nelson and Serafin1997). Mapping areas for conservation may be unable to protect species and ecosystems from broader environmental challenges, such as global warming or extreme weather events (Noss Reference Noss2001). By contrast, boundaries can also be overly effective, for instance severing ecological flows (Puth & Wilson Reference Puth and Wilson2001) or creating deleterious consequences at edges (Moenting & Morris Reference Moenting and Morris2006). Over time, conservation boundaries may come to have markedly different vegetation characteristics on either side.

Adding to these ideas, our focus on cartogeographic contributions suggests that conservation mappings can also be overly ‘effective’ at more abstract levels. For instance, mapping areas for conservation can implicitly suggest that relationships between species, ecosystems and territory are fixed in space, and will remain so into the future, contributing to static understandings of the environment that counter the inherent dynamism and flux of ecological conditions and conservation requirements (Zimmerer Reference Zimmerer2000). While ecologists are already aware of the dangers of this misapprehension, such misunderstandings of an overly-fixed relationship between conservation and space can be extremely influential in bureaucratic and popular domains. Thus mapped locations of forests in Thailand came, over time, to legally define ‘forests’ for state agencies, regardless of the actual vegetation found in those regions (Vandergeest Reference Vandergeest1996). In more popular senses, there is also the possibility that mapping an area as ‘wildlife reserve’ may create a false sense of security concerning the ability to achieve conservation goals, even possibly giving the impression that conservation has already happened and is successful. The abundance of ‘paper parks,’ which lack sufficient institutional capacity or funding to support their protected status and so ‘enjoy only a virtual existence as lines drawn on official maps’ (Terborgh & van Schaik Reference Terborgh, van Schaik, Terborgh, van Schaik, Davenport and Rao2002, p. 4), offers further evidence of such concerns.

A second issue of concern is that through designation of particular areas as PAs, other areas are concurrently implied, albeit unwittingly, to be less worthy of protection (Boothby Reference Boothby2002). Designation of conservation zones can therefore be taken to justify human use, and over-use, of areas outside the boundaries (Zimmerer Reference Zimmerer2000, p. 362). Here, the case of Costa Rica is instructive. While Costa Rica is proportionally well protected compared with other Latin American countries, in the 1990s it also experienced among the highest rates of deforestation owing to rapid depletion of forests outside these PAs (Vandermeer & Perfecto Reference Vandermeer and Perfecto1995). Clapp (Reference Clapp2004) suggested that PAs need not be viewed as bounded spaces with everywhere else defined as ‘unnatural.’ Instead, they should be considered core areas around which natural resources must be managed effectively across broad geographic ranges, reflecting discussions of continental-scale conservation.

The mapping of conservation spaces can also alter human perceptions of an area, creating new social and psychological associations. For instance, when Yellowstone National Park was founded in 1872, long-term human use of the land was effectively erased when Native American communities were displaced in order to create the Park. Similarly, the founding of Great Smoky Mountains National Park involved the removal of pioneer settlements to comply with visions of pristine nature. The ‘myth’ of wilderness that informed these actions was central to the creation of US national parks and has helped to perpetuate the idea of people as separate from nature (Cronon Reference Cronon and Cronon1995; Spence Reference Spence1999). Indeed, the mapping of a certain space as ‘national park’ or ‘nature reserve’ has, in many countries, traditionally been taken as implicit endorsement of the need to separate humans from nature; this is a clear contrast with current efforts to better integrate local populations and conservation spaces.

Conservation maps, as other maps, are necessarily reflective of and productive of power (see Harley Reference Harley1989). Just as cartographers necessarily privilege compass direction, area or other aspects in creating maps (Monmonier Reference Monmonier1996), those who engage in conservation mapping necessarily privilege certain species or understandings of nature over others. Further, the technological and scientific aspects of cartography have great political weight, at times encouraging the use of cartographic techniques over other approaches or forms of knowledge. Given the limitations of territorially-focused conservation, it is crucial to recognize the degree to which mapping, and computer-assisted technologies such as GIS in particular, may serve to formalize and legitimize territorial approaches to conservation, with the concurrent potential to sideline other non-territorial strategies. It is also noteworthy that many features of critical importance to the success of PAs are relatively ‘unmappable’ (for example social and political infrastructure) and may be neglected in a political climate that overemphasizes what can be achieved with innovative technological tools. Greater recognition of these issues might encourage more transparent and explicit decision-making with respect to conservation goals. Just as mapmakers must select particular cartographic projections that favour area, size or shape depending on their goals, so conservation practitioners might similarly be explicit with respect to which conservation goals are prioritized or likely to be met with particular conservation mappings. This would help to counter the false impression that all elements of a system can be preserved equally well under the PA banner.

Systematic and ecocentric conservation planning

Recognizing the degree to which protected areas are often designated and managed according to shifting notions of appropriate or desirable nature, as well as other social and political goals, calls have intensified for more ecocentered and systematic approaches to conservation planning (for example Theberge Reference Theberge1989; Woodley Reference Woodley, Nelson and Serafin1997; Simberloff Reference Simberloff1998; Margules & Pressey Reference Margules and Pressey2000). However, even when considering only ecological goals, prioritization is far from straightforward. Not only are there complex and contradictory challenges in meeting the needs of only non-human components of ecosystems (Schwartz Reference Schwartz1994), but PA outcomes also vary significantly depending on the priorities used (Whittaker et al. Reference Whittaker, Araújo, Jepson, Ladle, Watson and Willis2005). For instance, should representation of as many different ecosystems as possible in an array of PAs be the goal (Ricketts et al. Reference Ricketts, Dinerstein, Olson, Loucks, Eichbaum, DellaSala, Kavanagh, Hedao, Hurley, Carney, Abell and Walters1999), or is it more expedient to focus attention on biodiversity hotspots (Myers et al. Reference Myers, Mittermeier, Mittermeier, Fonseca and Kent2000)? An extensive literature has now focused on priority-setting in PA designation (Johnson Reference Johnson1995).

Existing PA systems are biased towards protecting particular types of ecosystems, often those in remote areas that are unsuitable for commercial activities (Margules & Pressey Reference Margules and Pressey2000). There tends to be bias towards the research and conservation of charismatic or better known forms of biodiversity (Ferrier et al. Reference Ferrier, Powell, Richardson, Manion, Overton, Allnutt, Cameron, Mantle, Burgess, Faith, Lamoreux, Kier, Hijmans, Funk, Cassis, Fisher, Flemons, Lees, Lovett and Van Rompaey2004; Seddon et al. Reference Seddon, Soorae and Launay2005), at the expense of other important aspects of an ecosystem that may face even greater threats. Furthermore, climate change and habitat loss are likely to make biases in terms of what is protected increasingly extreme (Pyke & Fischer Reference Pyke and Fischer2005). The notion of systematic and ecocentric approaches for locating and designing PAs has thus gained force. Systematic conservation assessment aims to use objective measures and PA selection algorithms to assess the relative merits of different candidate areas with respect to their ecological value (Margules & Pressey Reference Margules and Pressey2000; Pressey & Cowling Reference Pressey and Cowling2001; Moore et al. Reference Moore, Balmford, Allnutt and Burgess2004). Such approaches are often computer-based but may be coupled with procedures to try to develop sound implementation strategies and stakeholder collaboration (Knight et al. Reference Knight, Driver, Cowling, Maze, Desmet, Lombard, Rouget, Botha, Boshoff, Castley, Goodman, MacKinnon, Pierce, Sims-Castley, Stewart and Von Hase2006). These efforts move beyond traditional single species-focused conservation approaches, to recognition that biodiversity more broadly is a critical variable (Groves et al. Reference Groves, Jensen, Valutis, Redford, Shaffer, Scott, Baumgartner, Higgins, Beck and Anderson2002), underscoring the need for regional, rather than local, goal-setting. Assessing the complex notion of ‘biodiversity’ has emerged as a corollary challenge in efforts to prioritize PAs, with attempts to identify measurable proxies that effectively summarize overall ecological diversity (for example Andelman & Fagan Reference Andelman and Fagan2000; Hess & King Reference Hess and King2002; Lombard et al. Reference Lombard, Cowling, Pressey and Rebelo2003; Coppolillo et al. Reference Coppolillo, Gomez, Maisels and Wallace2004; Cowling et al. Reference Cowling, Knight, Faith, Ferrier, Lombard, Driver, Rouget, Maze and Desmet2004).

People and protected area management

In the conservation sciences there is also increasing attention to the need to focus on the interface between people's needs and interests and PA design and management. This integration is viewed by many as essential both to the promotion of human well-being and to ensuring that biodiversity and conservation needs are met in the long term (Sinclair et al. Reference Sinclair, Ludwig and Clark2000; Shafer Reference Shafer1999), as ‘declining biodiversity and the poverty and marginalization of remote communities are connected’ (Sheil et al. Reference Sheil, Puri, Wan, Basuki, van Heist, Liswanti, Rachmatika and Samsoedin2006, p.17). Attention along these lines has focused on both of these imperatives, namely the importance of resources and access to PAs for local populations and the critical role of socio-political factors in the success of PAs (see O'Connor et al. Reference O'Connor, Marvier and Kareiva2003).

Although there remains considerable controversy within the conservation community about how and if people should be reincorporated into PAs (for example Brandon et al. Reference Brandon, Redford, Sanderson, Brandon, Redford and Sanderson1998; Terborgh & van Schaik Reference Terborgh, van Schaik, Terborgh, van Schaik, Davenport and Rao2002), as well as concerns that expecting conservation to meet social and economic equity goals as well as ecological ones may be unreasonable and overly ambitious (Naughton-Treves et al. Reference Naughton-Treves, Holland and Brandon2005), there is growing awareness that PAs are nonetheless embedded in human-dominated landscapes (Zander & Kachele Reference Zander and Kachele1999; Curtin Reference Curtin2002; Sanderson et al. Reference Sanderson, Redford, Vedder, Coppolillo and Ward2002). Indeed, conservation efforts may be even more important in densely-populated areas considering the large numbers of threats in these places (Ricketts & Imhoff Reference Ricketts and Imhoff2003). With strong traditions of studying human aspects of resource use and people-park management in political ecology and human-environment geography, such issues provide clear opportunities for fruitful interdisciplinary engagement.

Protected area effectiveness and design

Owing to the many challenges of a limited network of PAs, and the particular challenges that come with separating activities within and outside them, mapping for conservation poses key dilemmas. Most notably, the goal is to preserve natural systems, yet the imposition of boundaries may disrupt processes that are critical to the evolution and maintenance of biological diversity. A wide variety of strategies have been suggested to improve the form and function of PAs. Strategies proposed have included efforts to reduce the negative effects of habitat fragmentation via concepts such as wildlife corridors and buffer zones (Beier and Noss Reference Beier and Noss1998; Shafer Reference Shafer1999; Berger Reference Berger2004). There is also increasing attention to moving beyond the limited and fragmented network of PAs towards management at regional or continental scales (Ricketts et al. Reference Ricketts, Dinerstein, Olson, Loucks, Eichbaum, DellaSala, Kavanagh, Hedao, Hurley, Carney, Abell and Walters1999; Groves et al. Reference Groves, Valutis, Vosick, Neely, Wheaton, Touval and Runnels2000) in response to evidence that PAs are likely to lose species over time in the absence of extensive and varied ranges (Newmark Reference Newmark1995). More specifically, emphasis has been placed on landscape permeability, connectivity and flexible boundaries, as well as on the creation of transboundary PAs that minimize the impacts of arbitrary international frontiers (for example see Danby Reference Danby1997; Noss Reference Noss2001; Hanks Reference Hanks2000; Roberts et al. Reference Roberts, Andelman, Branch, Bustamante, Castilla, Dugan, Halpern, Lafferty, Leslie, Lubchenco, McArdle, Possingham, Ruckelshaus and Warner2003; Berger Reference Berger2004).

In addition to such spatial approaches, there are also strategies to engender greater temporal flexibility in PA design and management. In particular, considerable work has addressed the challenge of making PAs more adaptive with respect to large-scale or long-term environmental changes, particularly climate change (for example Hulme Reference Hulme2005; Root & Schneider Reference Root and Schneider2006) although many of these approaches are, as yet, largely untested (Halpin Reference Halpin1997; Soto Reference Soto2001). Given that species are unlikely to remain ‘in place’ pending climate change, PA networks could nonetheless remain effective into the future if designed with this in mind (Hannah et al. Reference Hannah, Midgley, Andelman, Araújo, Hughes, Martinez-Meyer, Pearson and Williams2007). To this end, species range-shifts can be modelled and PA selection algorithms used to assess the effectiveness of PAs under particular environmental scenarios (Pressey & Cowling Reference Pressey and Cowling2001; Hannah et al. Reference Hannah, Midgley, Andelman, Araújo, Hughes, Martinez-Meyer, Pearson and Williams2007). All of these efforts constitute a mounting response to the limitations of fragmented and overly-fixed conservation mappings.

Strategies that upset eco-social inequalities

In line with efforts to improve equality in ecological senses via more systematic selection of PAs, mapping and its associated tools are being increasingly used to assess representativeness of coverage (Olson & Dinerstein Reference Olson and Dinerstein1998; Hazen & Anthamatten Reference Hazen and Anthamatten2004). GIS is also commonly used to identify inadequacies in PA systems at national and sub-national levels via gap analysis (Savitsky Reference Savitsky, Savitsky and Lacher1998). In other cases, GIS techniques have been used to assess equality of conservation coverage in broader senses, for instance, the degree to which global conservation priority maps guide the spending of the large conservation organizations (Halpern et al. Reference Halpern, Pyke, Fox, Haney, Schlaepfer and Zaradic2006), and whether sampling bias skews which areas are prioritized for conservation (Reddy & Dávalos Reference Reddy and Dávalos2003).

Geographic approaches have also been used to try to reduce inequalities in social and political senses. As noted, many theorists suggest that meeting the needs of local populations may actually improve the ability to meet conservation goals over the long term, for instance through minimizing unregulated access (Turner Reference Turner1999), or by achieving more robust and democratic resource management mechanisms (Agarwal Reference Agarwal2001). In response to such ideas, a complex of approaches has emerged that attempt to soften the divide between PAs and the needs of local populations, including: ‘multiple-use conservation areas’, ‘integrated conservation development projects’ and ‘nature-society hybrids’ (Zimmerer Reference Zimmerer2000; Clapp Reference Clapp2004). These ideals of integrating local interests and conservation needs have not gone unquestioned however, as goals of local people and conservationists do not necessarily intersect (Brandon et al. Reference Brandon, Redford, Sanderson, Brandon, Redford and Sanderson1998). Although PAs will never be able to achieve all social and ecological goals (Naughton-Treves et al. Reference Naughton-Treves, Holland and Brandon2005), discussions and efforts along these lines must proceed if protected areas as exclusionary fortresses are to be avoided (see Sunderlin et al. Reference Sunderlin, Angelsen, Belcher, Burgers, Nasi, Santoso and Wunder2005 for further discussion of integrating livelihoods and conservation needs).

There are also considerable efforts underway focused on using cartographic tools and techniques to challenge social and political inequalities (at times referred to as ‘counter-mapping’; Harris & Hazen Reference Harris and Hazen2006). For example, democratized mapping techniques or similar methods can be used to bolster resource access for indigenous groups or economically vulnerable populations (Nietschmann Reference Nietschmann1995; Maya People of Southern Belize 1997). Other examples demonstrate how participatory mapping technologies and skills, or use of basic sketch maps, can be used to foster greater participation of local people in conservation efforts (Eghenter Reference Eghenter2000; Hodgson & Schroeder Reference Hodgson and Schroeder2002; L. Pintea, personal communication 2002). The specific ways that local knowledge can be integrated in technical GIS environments, particularly for biodiversity aims, has also been explored (Anuchiracheeva et al. Reference Anuchiracheeva, Demaine, Shivakoti and Ruddle2003; Balram et al. Reference Balram, Dragicevic and Meredith2004). There is an impressive array of studies detailing use of mapping, participatory GIS and other advances for forest, marine, watershed and other conservation challenges (Convis Reference Convis2001). Attempts to promote popular access to powerful and persuasive cartographic tools raise concerns with respect to exploitation of local people, and questions of data quality and robustness (Hodgson & Schroeder Reference Hodgson and Schroeder2002). With serious attention to these concerns, however, there is still clear potential to extend such possibilities (Nietschmann Reference Nietschmann1995, p. 37).

Overcoming limitations of conservation mapping: engaging cartographic and GIS approaches

There are many additional ways in which cartogeographic and GIS approaches might be engaged more fully to further conservation goals. At a basic level, cartographers could contribute towards public knowledge campaigns through the production of visual representations to relay information about the urgency of species loss, drawing increased attention to conservation imperatives. GIS is already being widely used to focus conservation efforts by overlaying species' habitat requirements with information about climate, soil, human threats, social goals or other conditions to highlight priorities (for example Mazzotta et al. Reference Mazzotta, Magnuson, Opaluch and Johnston2002; Harris et al. Reference Harris, Jenkins and Pimm2005; Tchouto et al. Reference Tchouto, Yemefack, De Boer, De Wilde, Van Der Maesen and Cleef2006). There have been efforts to use GIS and spatial modelling techniques to evaluate landscape permeability, migration pathways and habitat suitability for particular species (Singleton et al. Reference Singleton, Gaines and Lehmkuhl2002; Ray et al. Reference Ray, Lehmann and Joly2002). GIS tools, in combination with increased computational power for statistical analysis, are also allowing more rigorous numerical analyses of data on species distributions than ever before (Tchouto et al. Reference Tchouto, Yemefack, De Boer, De Wilde, Van Der Maesen and Cleef2006). Use of GIS in these ways already enables integration of complex biophysical species requirements and could be extended to further incorporate data on human activities that are influential in conservation outcomes.

Real-time or time-series GIS data, modelling, and animated visualization techniques also hold potential, particularly to facilitate more flexible and dynamic conservation approaches. With such tools, the movements of certain protected populations can be monitored and the activities allowed within a certain area modified accordingly. For example, if a conservation effort is focused on chimpanzee populations, cartographic tools might enable conservation strategies to be adapted dynamically to chimpanzee troop movements, changing habitat characteristics or even fluctuating threats to chimpanzee populations. Applied to fisheries, new computer and mapping technologies have enabled commercial fishing fleets to monitor, track and harvest fish populations more effectively. Similar technologies may now be engaged to ensure the protection of fisheries by monitoring school movement or assessing marine conditions to identify productive areas to be able to enforce non-harvest (Valavanis et al. Reference Valavanis, Kapantagakis, Katara and Palialexis2004).

Emerging cartographic and visualization technologies might also be employed more extensively to engender greater seasonal flexibility, with regulations relaxed in seasons or years when conditions are particularly favourable and tightened under stressful conditions. For example, work on spatio-temporal variability in productivity of natural systems (for example Valavanis et al. Reference Valavanis, Kapantagakis, Katara and Palialexis2004) could be used to guide changing seasonal quotas and regulations. Such seasonal flexibility is already endorsed by a number of management authorities. Thus the seasonally flexible boundary of the Royal Bardia National Park (Nepal) allows villagers periodic access to collect plant products for their own use (Brown Reference Brown1997). Similar suggestions have been made to allow seasonal passage of herders through PA territories (Turner Reference Turner1999) in order to mesh environmental management practices with social justice goals. Fuzzy boundaries and animated mapping could also help conceptualize and develop boundaries that are more flexible, perhaps more consistent with buffer zones and similar notions.

Advances in geography and cartography are opening up new possibilities in terms of the scale at which territorially-focused conservation takes place, moving away from the national and sub-national scales at which decisions have traditionally been made. Computer-assisted mapping technologies have already facilitated a move from local- to regional- and global-scale analyses, which has enabled the compilation of a variety of global-scale spatial databases (Olson et al. Reference Olson, Dinserstein, Wikramanayake, Burgess, Powell, Underwood, Itoua, Strand, Morrison, Loucks, Allnutt, Ricketts, Kura, Lamoreux, Wettengel, Hedao and Kassem2001; WDPA Consortium 2006). Independent researchers, too, are increasingly thinking through other scales that might be appropriate to particular circumstances (for example Turner Reference Turner and Zimmerer2006).

Overcoming limitations of the contemporary patchwork of protected areas

There are also more radical possibilities suggested by our geographic discussion of limits to territorially-focused conservation. Some of the critiques raised above invite the exploration of possibilities beyond the very idea of the PA. For instance, mapping and GIS technologies have helped to reveal the extent of many animals' ranges and their complex use of different ecosystems, highlighting the need for extensive areas over which certain species can range and the need for connections between ecosystems. This, coupled with greater understanding of the limitations of territorial approaches, could foster discussion of how to focus attention beyond PAs and on to other landscapes, for instance, urban and agricultural landscapes that represent a large proportion of the Earth's surface. Attention could very fruitfully be paid to improving conditions in these partially-degraded environments (Ricketts & Imhoff Reference Ricketts and Imhoff2003; Lefroy & Smith Reference Lefroy and Smith2004) through moderating human activity and consumption, rather than focusing heavily on ‘wilderness areas’ (Cronon Reference Cronon and Cronon1995). Such strategies are consistent with a wide variety of schema underway, including clean air legislation, neighbourhood parks, environmental education and movements such as bioregionalism (McGinnis Reference McGinnis1999). This approach was also used by the European Union's Habitats Directive of 1992, which aimed to incorporate human activities into conservation approaches, replacing a previous policy that aimed to set nature aside from human activities (Alphandéry & Fortier Reference Alphandéry and Fortier2001). Thus future conservation initiatives must necessarily move beyond simple legal delineation of wildlife areas towards depending more on fostering stakeholder agreements (Boothby Reference Boothby2002, p. 45). In all these ways, conservation can be extended beyond the limited idea of the ‘protected area.’